Autho

r's

pers

onal

co

py

Fluorinated and hemifluorinated surfactants derived from maltose:Synthesis and application to handling membrane

proteins in aqueous solution

Ange Polidori,a,* Marc Presset,a Florence Lebaupain,b Bruno Ameduri,c Jean-Luc Popot,b

Cecile Breytonb,* and Bernard Puccia,*

aUniversite d’Avignon et des Pays du Vaucluse, Laboratoire de Chimie Bioorganique et des Systemes Moleculaires Vectoriels,

33 rue Louis Pasteur, F-84000 Avignon, FrancebLaboratoire de Physico-Chimie Moleculaire des Membranes Biologiques, UMR 7099, CNRS and Universite Paris-7,

Institut de Biologie Physico-Chimique CNRS FRC 550, 13 rue Pierre et Marie Curie, F-75005 Paris, FrancecLaboratoire de Chimie Macromoleculaire, ENSCM, 8, rue de l’Ecole Normale, F-34296 Montpellier Cedex 5, France

Received 13 July 2006; revised 13 August 2006; accepted 14 August 2006

Available online 8 September 2006

Abstract—Two novel fluorinated surfactants have been obtained by grafting by radical reaction either a fluorocarbon or an ethylend-capped fluorocarbon chain onto the double bond of b-DD-allyl maltose. The two compounds thus obtained form polydisperseaggregates in water. They can keep membrane proteins water-soluble, but the protein/surfactant complexes are polydisperse, whichaffects neither the native state nor the stability of the proteins.� 2006 Elsevier Ltd. All rights reserved.

Obtaining aqueous solutions of membrane proteins(MPs) so as to study their function and structure is ofmajor scientific and biomedical importance. MPs repre-sent 20–40% of open-reading frames in fully sequencedgenomes.1 They fulfil a wide variety of functions rangingfrom signal reception and transduction to solute translo-cation and are the target of more than half of pharma-ceutical drugs. However, studying them is difficult, asreflected in the fact that <1% of current PDB entries de-scribe membrane protein structures. Several bottlenecksexplain this situation, among which the inactivating ef-fect of the detergents that are used to extract MPs fromtheir native membrane environment for the purpose ofin vitro biochemical, functional and structural charac-terisation.2 Detergents substitute for lipids around thehydrophobic domain of membrane proteins, formingwater-soluble complexes. However, the dissociating ef-fect of detergents, which allows them to solubilise bio-logical membranes, can be difficult to control, resulting

in the destabilisation and irreversible inactivation ofsolubilised MPs.2 Two likely mechanisms leading toinactivation are the intrusion of the detergent into thetransmembrane region of the protein and the dissocia-tion of stabilising lipids, cofactors or subunits.2,3

To alleviate this problem, less ‘aggressive’ surfactantsare being developed.4,5 Non-ionic fluorinated surfac-tants6,7 are lipophobic and, as such, do not solubilisebiological membranes.7,8 They are, however, able tokeep water-soluble MPs that have been extracted usingclassical hydrogenated surfactants.7 In order to improvetheir affinity for the hydrophobic, hydrocarbon-liketransmembrane surface of MPs, we have introducedthe so-called ‘hemifluorinated’ surfactants, which bearan ethyl hydrogenated tip at the extremity of the fluori-nated chain (Fig. 1).9–12

The first hemifluorinated surfactant to be synthesised,9

HF-TAC (Fig. 1), proved able to keep soluble a set oftest MPs in their native and functional state, while sta-bilising them as compared to classical detergents usedat comparable concentrations.11 The large and highlysoluble polar head of HF-TAC was chosen in order tocounterbalance its highly insoluble hemifluorinated tail.

0960-894X/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.bmcl.2006.08.070

Keywords: Fluorinated surfactants; Maltoside; Membrane proteins;

Solubilization; Aggregates.* Corresponding authors. Tel.: +33490144442; fax: +33490144449;

e-mail addresses: bernard.pucci@univ-avignon.fr; cecile.breyton@

ibpc.fr

Bioorganic & Medicinal Chemistry Letters 16 (2006) 5827–5831

Autho

r's

pers

onal

co

py

While convenient to test the concept of hemifluorinatedsurfactants, HF-TAC suffers from a serious drawbackfrom the point of view of biochemical and structuralapplications: because the synthesis of its polar head in-volves radical polymerisation, it is inevitably polydis-perse, which can lead to batch-to-batch variations. Totry to circumvent this problem, we have undertaken toprepare hemifluorocarbon surfactants bearing monodis-perse polar heads (Fig. 1). The zwitterionic aminoxydemoiety of HF-AO,10 however, was found to decreaseMP stability. Since several very useful detergents inMP biochemistry, such as b-DD-octylglucoside or b-DD-do-decylmaltoside,2 bear a glycosidic head, we focused onthe synthesis of glycoside-derived fluorinated surfac-tants. In a first step, we synthesised surfactants with alactobionamide head group (HF-Lac, Fig. 1).12 HF-Lac was found to be efficient in keeping several testMPs water-soluble, active and monodisperse.12 The lac-tobionamide polar head, however, proved destabilisingas compared to the maltoside polar head of DDM, asevidenced by the inactivating character of the hydroge-

nated homologue of HF-Lac.12 The work presentedherein reports on the synthesis, physical–chemical andbiochemical characterisation of fluorinated (F-Malt)and hemifluorinated (HF-Malt) homologues of thewidely used detergent b-DD-dodecylmaltoside (DDM).

The most efficient method to introduce an alkyl chain inthe b anomeric position of maltose consists in (i) increas-ing the reactivity of the anomeric carbon 1 of a peracet-ylated maltose by grafting it with trichloroacetimidate,and (ii) substituting this group with an alcohol.13 Thisreaction leads exclusively to the b-alkoxymaltoside iso-mer, generally in good yield. However, fluorocarbonalcohols exhibit a lower nucleophilic power than theirhydrocarbon analogues, due to the electron-withdraw-ing effects of fluorine atoms. The coupling assays we per-formed with 3,3,4,4,5,5,6,6,7,7,8,8-dodecafluorooctanoland trichloroacetimidate maltoside proved negative.14

To graft a fluorinated chain on the carbon anomer ofdifferent glycosides, Bonnet-Delpon et al.15 used theMitsunobu reaction. Another approach developed byMietchen et al. consists in grafting a perfluoroalkyl io-dide onto an allyl glycoside in the presence of a radicalinitiator.16 We chose this chemical pathway. First, theallyl glycoside 1 (Scheme 1) was obtained by condensa-tion of allyl alcohol on peracetylated maltoside in thepresence of a Lewis acid.17 Amongst the different Lewisacids assayed as catalyst, BF3–etherate gave the highestcoupling yield (89%). This reaction is accompanied by apartial deacetylation of the maltose derivative 1, due toa transesterification reaction with the allyl alcohol usedin excess. To minimise the effect of this side reaction, onecan either use only 1.5 equiv of the allyl alcohol or, pref-erably, perform a peracetylation of crude products ofthe reaction in the presence of acetic anhydride/pyridine,followed by purification by silica gel chromatography.As regards the hemifluorocarbon diiodide 2, it was syn-thesised by radical addition of diiodoperfluorohexaneon ethylene following the method already reported byManseri et al.18 Condensation of compound 2 on allyl

FF

FF

FF

FF

FF

FFS

H

NH

OH OH

OHO

7-8

FF

FF

FF

FF

FF

FF

NN

OO

FF

FF

FF

FF

FF

FF HN

O

HO

OHO

OHHO

O OH

OHHO

HO

HF-TAC

HF-AO

HF-Lac

Figure 1. First-generation hemifluorinated non-ionic surfactants.9–12

AcO OAcO

AcO

OAc

O OAcO

OAcAcO

OAc

II

F F

F F

F F

F F

F F

F F

AcO O

AcOAcO

OAc

O OAcO

AcO

OAc

O

IF F

F F

F F

F F

F F

F FI

FF F

F F

F F

F F

F F

F F

O

HOO

HOHO

OH

OO

HOHO

OH

F F

F F

F F

F F

F F

F F

OI

AcO OAcO

AcO

OAc

O OAcO

AcO

OAc

I

O

HOO

HOOH

OH

O OHO

OH

OHF F

F F

F F

F F

F F

F F

HF-Malt 4F-Malt 5

d,e

a

b

c1

2

3

Scheme 1. Synthesis of HF-Malt. Reagents and conditions: (a) allyl alcohol, BF3/Et2O, DCM, argon, 89% yield; (b) ethylene, CuI, 165 �C, 80%

yield;18 (c) Zn dust, DCM, argon, 61–88% yield; (d) Bu3SnH, AIBN, acetonitrile, argon, 67% yield; (e) MeONa, MeOH, 99% yield.

5828 A. Polidori et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5827–5831

Autho

r's

pers

onal

co

py

maltoside derivative 1 was carried out in the presence ofZn powder in methylene choride to provide compound 3in good yield (66%).19 Iodide groups were then reducedusing tributyltin hydride (Bu3SnH) in the presence ofa,a 0 azo-bis-isobutyronitrile (AIBN), yielding hemiflu-orocarbon peracetylated maltoside. The observed par-tial deacetylation of the maltose moiety was againcorrected by acetylation of crude products. After purifi-cation by silica gel chromatography, peracetylated hemi-fluoro-maltoside was isolated with 67% yield, and,finally, quantitatively deacetylated at room temperaturein methanol in the presence of catalytic amount of sodi-um methylate following the Zemplen method. After finalpurification on a Sephadex LH20 column, HF-Malt 4was isolated as a white powder and characterised byNMR.20 The same procedure was applied to perflu-orohexyl iodide to provide F-Malt (compound 5) in52% overall yield.

F- and HF-Malt are soluble in water. Physical–chemicalparameters derived from tensiometric measurementswere compared to those of their hydrogenated homo-logue DDM (Table 1).

As previously observed both for telomeric surfactantsderived from Tris (HF-TAC and F-TAC)9 and for lacto-bionamide derivatives (HF-Lac and F-Lac),12 adding anethyl group at the end of the fluorocarbon chain has lit-tle effect on the critical micellar concentration (CMC)(Table 1). Furthermore, whereas the limit surface ten-sion of F-Malt is lower than that of DDM, as expectedfor fluorinated compounds,21 that of HF-Malt has anintermediate value. The area per molecule of surfactantis higher for HF-Malt than for either DDM or F-Malt, aphenomenon already noted for HF-Lac.12 These unusu-al features can be explained by a poorer packing be-tween fluorinated and hydrogenated groups, by sterichindrance in the core of HF-Lac micelles and/or bythe acidic character of the methylene group vicinal to

the fluorocarbon chain.12 Both the fluorinated chainand the presence of the ethyl tip at its extremity also af-fect the aggregation behaviour of F- and HF–Malt(Table 2): whereas DDM forms small, spherical micelles(�6 nm in diameter),22 F-Malt, at 25 �C and at relativelylow concentration (0.4 mM), forms larger aggregates(Table 2)—probably cylindrical assemblies, as do otherfluorinated surfactants.23 HF-Malt forms even largerand more polydisperse aggregates (Table 2). This behav-iour is enhanced at higher concentration (5 mM) as wellas at lower temperature (4 �C), conditions under whicha population of micrometric particles is observed(Table 2).

The ability of F- and HF-Malt to keep MPs water-solu-ble was investigated using three model proteins repre-sentative of different types of structures: (i)bacteriorhodopsin (BR), an archaebacterial proteinfolded into a bundle of seven transmembrane a-heliceswith small extramembrane loops.24 A molecule of retinalis covalently but loosely bound to the protein, whosevisible absorption spectrum is a sensitive and convenientreporter of whether it is in its native state or not; (ii)tOmpA, an eight-strand b-barrel that forms the trans-membrane region of outer membrane protein A fromthe Gram-negative eubacterium Escherichia coli;25 and(iii) cytochrome b6 f, a photosynthetic complex extractedfrom the chloroplast of Chlamydomonas reinhardtii.Cytochrome b6 f is a superdimer, each monomer beingcomprised of eight transmembrane subunits and numer-ous cofactors, including stabilising lipids.26,27 Its stabil-ity can be monitored enzymatically.26 All threeproteins were purified using classical detergents.27–29

Figure 2 shows the migration of BR in 10–30% sucrosegradients (as described in Ref. 11). This procedure en-ables surfactant exchange, while the position of the pro-tein/surfactant band yields an estimate of thesedimentation coefficient of the particles (which is relat-ed to their mass, size, shape and density) and its widthan indication of their mono- or polydispersity. InDDM, BR migrates as a sharp band, in the upper partof the gradient. When transferred to either F- or HF-Malt, it migrates deeper. A similar behaviour was previ-ously observed with other fluorinated surfactants,11,12

and can be explained by the high density conferred tothe surfactants by the fluorine atoms. The bands are alsomuch broader in F- or HF-Malt than in detergent,which reveals a high polydispersity of the MP/surfactant

Table 1. Physical–chemical properties of DDM, F-Malt and HF-Malt

Compound CMC (mM) ccmc (mN m�1) Area per

molecule (nm2)

DDM 0.17a 35.4a 0.42a

F-Malt 0.20 ± 0.01 17.5 ± 0.5 0.38 ± 0.02

HF-Malt 0.22 ± 0.01 27 ± 0.5 0.51 ± 0.03

a Values from Ref. 22.

Table 2. Size of the aggregates formed by DDM, F-Malt and HF-Malt

Compound 25 �C, 0.4 mMa 4 �C, 0.4 mMa 4 �C, 5 mMa

Diameter

(nm)

HHWb (nm) % (volume) Diameter (nm) HHWb (nm) % (volume) Diameter (nm) HHWb (nm) %

(volume)

DDM 6.3 ± 0.3 1.6 ± 0.1 100 7.2 ± 0.3 1.6 ± 0.1 100 8.2 ± 0.5 3.1 ± 0.5 100

F-Malt 11 ± 2 1.5 ± 0.3 99 20 ± 2 7 ± 1 97 25 ± 5 20 ± 10 99

HF-Malt 16 ± 2 4 ± 3 99 19 ± 5 12 ± 3 94 32 ± 20 21 ± 7 88

650 ± 330 200 ± 100 6 1000 ± 300 480 ± 200 12

a Micelle sizes measured by dynamic light scattering using a Zetasizer Nano-S model 1600 (Malvern Co). The sample was prepared 24 h before the

measurement and filtered through a 0.45-lm filter just before measurements.b HHW, the width of the peak at half-height, an indication of the degree of polydispersity of the aggregates.

A. Polidori et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5827–5831 5829

Autho

r's

pers

onal

co

pycomplexes. Polydispersity can stem from either compo-nent of the complex: the protein, which can form small,heterogeneous aggregates, or the surfactant, differentamount of which can bind to the protein. BR migratesat the same position in F- or HF-Malt gradients as inan HF-Lac gradient (Fig. 2), suggesting that the proteinhas not aggregated and is present under the same oligo-meric state in all conditions. On the other hand, dynam-ic light scattering measurements show that F- andHF-Malt form large and heterogeneous aggregates(Table 2). Thus, the high dispersity of BR/surfactantcomplexes seems more likely to stem from heteroge-neous surfactant binding, presumably as a consequenceof the tendency of the latter to form aggregates ofvariable size. The BR band in F-Malt gradients was ob-served to be slightly but reproducibly thinner than thatin HF-Malt gradients, consistent with the fact thatF-Malt forms by itself slightly less heterogeneous aggre-gates than does HF-Malt (Table 2).

The same overall behaviour was observed with the b6 fcomplex and with tOmpA: MP/surfactant complexesmigrated deeper into F- and HF-Malt gradients thaninto DDM ones, and the bands were much broader(not shown). The behaviour of MP/HF-Malt complexesis noticeably different from that of complexes formedwith either F- or HF-Lac.12 F- and HF-Lac also formby themselves aggregates of variable size.12 However,their complexes with MPs are monodisperse (Ref. 12and Fig. 2). This likely results from F- and HF-Lacaggregates being smaller,12 that is, featuring a strongerspontaneous interface curvature.

At low HF-Malt concentration (0.5 mM), tOmpA mi-grates near the bottom of the tube, indicating the onsetof aggregation. When transferred into increasing surfac-tant concentrations (2–10 mM), the protein/surfactantband shifts towards lighter species, consistent with pro-tein/surfactant interactions progressively overcomingprotein/protein interactions. However, the position ofthe band remained beneath that observed in HF-Lac,suggesting that, even at 10 mM HF-Malt, tOmpA formssmall aggregates (not shown).

The integrity of BR and of the b6 f complex in DDM,F- or HF-Malt was investigated by spectroscopic and/or enzymatic measurements. Notwithstanding the poly-dispersity of the complexes, BR and the b6 f complexwere found to be in their native state, and the b6 f com-plex enzymatically as active as in DDM. Their stabilityover time was either comparable to (at low surfactant

concentrations) or higher than (at higher surfactant con-centrations) that in DDM (not shown), as previouslyobserved both with HF-TAC11 and HF-Lac.12

We have described the synthesis of novel fluorinated orhemifluorinated surfactants derived from maltose. Inkeeping with earlier observations,9,10,12 addition of anethyl tip to the fluorinated chain seems to destabilisethe micelles, possibly because of unfavourable interac-tions between fluorinated and hydrogenated groups.Unlike surfactants of the TAC series,30 however,F- and HF-Malt form large and polydisperse aggre-gates, as already observed for the lactobionamideseries.12 At variance with the latter case, however, theheterogeneity of (H)F-Malt aggregates seems to entailthat of their complexes with MPs (Fig. 2). Future workwill thus focus on optimising the chemical structure offluorinated and hemifluorinated surfactants with theaim of obtaining compounds that form small, globularaggregates. Upon comparing the chemical structuresand physical–chemical properties of compounds of the(H)F-TAC, (H)F-Lac and (H)F-Malt series, one is ledto hypothesise that the introduction of a bulkier polarhead should induce a decrease of the curvature radiusof the aggregates and thus favour the formation of mi-celle-like aggregates and monodisperse MP/surfactantcomplexes, while preserving the useful biochemicalproperties exhibited thus far by hemifluorinatedsurfactants.

Acknowledgments

This work was supported by the C.N.R.S., the C.E.A.,the Universities of ‘Paris-7 Denis Diderot’ and ‘Avignonet Pays du Vaucluse’, and by the E.C. Specific TargetedResearch Project Innovative tools for membrane struc-tural proteomics. F.L. and M.P. were the recipients offellowships from the ‘Ministere de l’EnseignementSuperieur et de la Recherche’.

References and notes

1. Wallin, E.; von Heijne, G. Protein Sci. 1998, 7, 1029.2. le Maire, M.; Champeil, P.; Møller, J. V. Biochim.

Biophys. Acta 2000, 1508, 86.3. Breyton, C.; Tribet, C.; Olive, J.; Dubacq, J.-P.; Popot, J.-

L. J. Biol. Chem. 1997, 272, 21892.4. Sanders, C. R.; KuhnHoffmann, A.; Gray, D. N.; Keyes,

M. H.; Ellis, C. D. ChemBioChem 2004, 5, 423.5. Gohon, Y.; Popot, J.-L. Curr. Opin. Colloid Interface Sci.

2003, 8, 15.6. Maurizis, J. C.; Pavia, A. A.; Pucci, B. Bioorg. Med.

Chem. Lett. 1993, 3, 161.7. Chabaud, E.; Barthelemy, P.; Mora, N.; Popot, J.-L.;

Pucci, B. Biochimie 1998, 80, 515.8. Barthelemy, P.; Tomao, V.; Selb, J.; Chaudier, Y.; Pucci,

B. Langmuir 2002, 18, 2557, and references cited.9. Barthelemy, P.; Ameduri, B.; Chabaud, E.; Popot, J.-L.;

Pucci, B. Org. Lett. 1999, 1, 1689.10. Chaudier, Y.; Zito, F.; Barthelemy, P.; Stroebel, D.;

Ameduri, B.; Popot, J.-L.; Pucci, B. Bioorg. Med. Chem.Lett. 2002, 12, 1587.

Figure 2. Migration of bacteriorhodopsin in 10–30% sucrose gradients

in the presence of either DDM, F- or HF-Malt (0.5 or 5 mM) or HF-

Lac (5 mM). Gradients were centrifuged 4 h at 200,000·g.

5830 A. Polidori et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5827–5831

Autho

r's

pers

onal

co

py

11. Breyton, C.; Chabaud, E.; Chaudier, Y.; Pucci, B.; Popot,J.-L. FEBS Lett. 2004, 564, 312.

12. Lebaupain, F.; Salvay, A. G.; Olivier, B.; Durand, G.;Fabiano, A. -S.; Michel, N.; Popot, J. -L.; Ebel, C.;Breyton, C.; Pucci, B. Langmuir., in press.

13. Yeung, B. K. S.; Adamski-Werner, S. L.; Bernard, J. B.;Poulenat, G.; Petillo, P. A. Org. Lett. 2000, 2, 3135.

14. Olivier, B. Thesis, Universite d’Avignon et des pays duVaucluse, December 2004.

15. Gueyrard, D.; Rollin, P.; Thanh Nga, T. T.; Ourevitch,M.; Begue, J.-P.; Bonnet-Delpon, D. Carbohydr. Res.1999, 318, 171.

16. Hein, M.; Miethchen, R. Eur. J. Org. Chem. 1999, 2429.17. Dahmen, J.; Magnusson, G. Carbohydr. Res. 1983, 114, 328.18. Manseri, A.; Ameduri, B.; Boutevin, B.; Kotora, M.;

Hajeck, M.; Caporicio, G. J. Fluor. Chem. 1995, 73, 151.19. Requirand, N.; Blancou, H.; Commeyras, A. Bull. Soc.

Chim. Fr. 1993, 130, 798.20. Spectra of F-Malt and HF-Malt are summarised as follows:

HF-Malt. Acetylated compound: Rf: 0.73 (ethyl acetate/Cy: 6/4), mp: 59 �C (dec), ½a�20

D : +32.4 (c, 1, DCM), 1H NMR(/CDCl3): d (ppm) 5.42 (1H, d, H 01, 3JH–H = 4 Hz), 5.36(1H,t, H3), 5.26 (1H, t, H 03), 5,06(1H, m, H 04), 4.86 (1H, m, H 02),4.79 (1H, m, H2), 4.53 (1H, d, H1, 3JH–H = 8 Hz), 4.45 (1H,m, H6), 4.23 (2H, m, 2� H 06), 3.97 (5H, m,H4 þ H6 þ H 05 þ CH2O), 3.61 (1H, m, H5), 2.07 (27H, m,7 · CH3CO + CH2CH2CF2 + CH2CH3), 1.14(3H, t,CH2CH3), 13C NMR (/CDCl3): d (ppm) 170.6 (CH3CO),170.5 (CH3CO), 170.3 (CH3CO), 170 (CH3CO), 169.9(CH3CO), 169.6 (CH3CO), 169.5 (CH3CO),130.0! 110.0 (6 · CF2), 100.2(C1), 95:5 ðC01Þ, 75.3(C4),72.6 (C3), 72.1 (C2), 72.0 (C5), 70:0 ðC03Þ, 69:3 ðC05Þ,68:5 ðC02Þ, 68.4 (CH2O), 68:0 ðC04Þ, 62:7 ðC06Þ, 61.4 (C6),27.9 (OCH2CH2CH2CF2), 27.6 (CH2CH2CF2), 24.6(CF2CH2CH3), 21.0 (CH3CO), 20.9 (CH3CO), 20.8(CH3CO), 20.7 (CH3CO), 20.6 (CH3CO), 20.5 (CH3CO),20.4 (CH3CO), 4.5 (CH2CH3), 19F NMR (/CDCl3): d (ppm)�114.45 (2F, m, CF2 (CH2)3O),�116.42 (2F, m,CF2CH2CH3), �121.91 (4F, m, CF2CF2(CF2)2CF2CF2),�123.71 (4F, m, CF2CF2 (CF2)2CF2CF2), Deacetylatedcompound 4: Rf: 0.62 (ethyl acetate/MeOH/H2O: 7/2/1),mp: 94 �C (dec), ½a�20

D : +37.2 (C, 1, DCM), 1H NMR(DMSO-d6): d (ppm) 5.52 (1H, m, OH), 5.46 (1H, m, OH),5.18 (1H, d, H 01, 3JH–H = 5 Hz), 5.02 (1H, m, OH), 4.92 (2H,m, 2 · OH), 4.52 (2H, m, 2 · OH), 4.20 (1H, d, H1, 3JH–H =8 Hz), 3.70! 3.03 (12H, m, H2 þ H3 þ H4 þ H5þ2� H6 þ H 02 þ H 03 þ H 04 þ H 05 þ 2� H 06þCH2O), 2.25(4H, m, 2 · CH2CF2), 1.78 (2H, m, OCH2-CH2CH2CF2),1.07 (3H, t, CH2CH3), 13C NMR (DMSO-d6): d (ppm)125.0! 105.0 (6 · CF2), 103.1 (C1), 101:3 ðC01Þ, 80.4 (C4),76.8 (C3), 75.6 (C2), 73.9 (C5), 73:8 ðC03Þ, 73:4 ðC05Þ, 72:9 ðC02Þ,70:3 ðC04Þ, 67.6 (CH2O), 61:2 ðC06Þ, 61.1 (C6), 27.4(OCH2CH2CH2CF2), 24.2 (CF2CH2CH3), 20.9(OCH2CH2CH2CF2), 4.8 (CH2CH3),19F NMR (DMSO-d6): d (ppm)�113.35 (2F, m, CF2(CH2)3O),�115.23 (2F, m,CF2CH2CH3), �121.73 (4F, m, CF2CF2(CF2)2CF2CF2),�123.14 (4F, m, CF2CF2 (CF2)2CF2CF2). F-Malt. Acety-

lated compound: Rf : 0.67 (ethyl acetate/cyclohexane: 6/4),mp: 68 �C (dec), ½a�20

D : +51.6 (c, 1, DCM), 1H NMR(CDCl3): d (ppm) 5.44 (1H, d, H01), 5.38! 5.24(2H, m,H3 þH03), 5.08 (1H, m, H04), 4.91! 4.81 (2H, m, H02 þH2),4.55 (1H, d, H1), 4.31! 4.40 (9H, m, H4 þH5 þ 2�H6 þH052�H06 þ CH2O), 2.05 (25H, m, 7 · CH3CO +CH2CF2 + OCH2-CH2CH2CF2), 13C NMR (CDCl3): d(ppm) 170.6 (CH3CO), 170.5 (CH3CO), 170.2 (CH3CO),170 (CH3CO), 169.9 (CH3CO), 169.6 (CH3CO), 169.4(CH3CO), 130.0! 110.0 (5 · CF2 + CF3), 100.2 (C1),95:6 ðC01Þ, 75.3 (C4), 72.7 (C3), 72.2 (C2), 72.1 (C5),70:0 ðC03Þ, 69:4 ðC05Þ, 68:5 ðC02Þ, 68.3 (CH2O),68:1 ðC04Þ, 67.0(OCH2CH2CH2CF2), 62:7 ðC06Þ, 61.5 (C6), 27.5 (CH2CF2),21.0 (CH3CO), 20.9 (CH3CO), 20.8 (CH3CO), 20.7(CH3CO), 20.6 (CH3CO), 20.5 (CH3CO), 20.4 (CH3CO),19F NMR (CDCl3): d (ppm) �80.73 (3F, m, CF3), �114.36(2F, m, CF2CH2), �121.86 (2F, m, CF2CF3), �122.82 (2F,m, CH2CF2CF2CF2), �123.46 (2F, m, CH2CF2CF2),�126.07 (2F, m, CF2CF2CF3). Deacetylated compound 5:Rf : 0.61 (ethyl acetate/MeOH/H2O: 7/2/1), ½a�20

D : +40.7 (c, 1,DCM), 1H NMR (/MeOH-d4): d (ppm) 5.18 (1H, m, OH),4.96 (1H, d, H 01), 4.90 (5H, m, 5 · OH), 4.31 (1H, d, H1),4.10! 3.20 (15H, m, H2 þ H3 þ H4 þ H5 þ 2� H6þH 02þH 03þH 04þH 05þ2�H 06þCH2OþOH ), 2.34 (2H, m,CH2CF2), 1.93 (2H, m, OCH2CH2CH2CF2), 13C NMR(MeOH-d4): d (ppm) 130.0! 110.0 (5 · CF2 + CF3), 102.8(C1), 101:5ðC01Þ, 79.9 (C4), 76,4(C3), 75,2(C2), 73.7 (C5),73:4ðC03Þ, 73:2ðC05Þ, 72:8ðC02Þ, 70:1ðC04Þ, 67.8(CH2O),61:3ðC06Þ, 60.8 (C6), 27.4 (OCH2CH2CH2CF2),20.5 (OCH2CH2CH2CF2), 19F NMR (/MeOH-d4): d(ppm) �82.39 (3F, m, CF3), �115.47 (2F, m, CF2CH2),�122.96 (2F, m, CF2CF3), �123.93 (2F, m,CH2CF2CF2CF2), �124.46 (2F, m, CH2CF2CF2),�127.34 (2F, m, CF2CF2CF3).

21. Kissa, E. In Fluorinated Surfactants: Synthesis, Properties,Applications. Surfactants Science Series 50; Dekker: NewYork, 1994; p 92, Chapter 4.

22. Hines, J. D.; Thomas, R. K.; Garrett, P. R.;Rennie, G. K.; Penfold, J. J. Phys. Chem. B 1997,101, 9215.

23. Kissa, E. In Fluorinated Surfactants: Synthesis, Properties,Applications. Surfactants Science Series 50; Dekker: NewYork, 1994; p 264, Chapter 7.

24. Pebay-Peyroula, E.; Rummel, G.; Rosenbusch, J. P.;Landau, E. M. Science 1997, 277, 1676.

25. Pautsch, A.; Schulz, G. E. Nat. Struct. Biol. 1998, 5, 1013.26. Pierre, Y.; Breyton, C.; Kramer, D.; Popot, J.-L. J. Biol.

Chem. 1995, 270, 29342.27. Stroebel, D.; Choquet, Y.; Popot, J.-L.; Picot, D. Nature

2003, 426, 413.28. Pautsch, A.; Vogt, J.; Model, K.; Siebold, C.; Schulz, G.

E. Proteins 1999, 34, 167.29. Oesterhelt, D.; Stoeckenius, W. Methods Enzymol. 1974,

31, 667.30. Dupont, A.; Eastoe, J.; Barthelemy, P.; Pucci, B.;

Heeman, R.; Penfold, J.; Steytler, D. C.; Grillo, I. J.Colloid. Interface Sci. 2003, 261, 184.

A. Polidori et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5827–5831 5831